This project involves evaluating common human and animal paramyxoviruses as potential human vaccine vectors against highly pathogenic viruses. We previously evaluated human parainfluenza virus type 3 (HPIV3) as a vector to express the spike glycoprotein of Severe Acute Respiratory Syndrome Coronavirus (SARS). A single dose of the HPIV3-S construct administered by the combined intranasal (IN) and intratracheal (IT) routes was immunogenic and protective against SARS challenge in African green monkeys (AGM). We also previously evaluated HPIV3 as a vector to express the single glycoprotein GP of Ebola virus (EBOV). This construct was highly immunogenic and completely protective in guinea pigs against an adapted strain of EBOV. A single IN/IT inoculation of rhesus monkeys was moderately immunogenic against EBOV and protected 88% of the animals against severe hemorrhagic fever and death caused by EBOV challenge. Two doses were highly immunogenic and all of the animals were free of disease signs and detectable EBOV challenge virus. Since HPIV3 is a common human pathogen and essentially all adults have a history of natural infection with HPIV3, it was important to determine whether previous infection with HPIV3 would restrict the replication and immunogenicity of the HPIV3 vector. In guinea pigs that were infected with HPIV3 and challenged 40 days later with HPIV3/EboGP, replication of the vector could not be detected, indicating a high level of restriction. Surprisingly, however, the immune response to EBOV GP was almost equivalent to that achieved in control animals that had not been previously infected with HPIV3. Next, rhesus monkeys were infected twice with HPIV3 and, 11 months following the second infection, were immunized with two doses of HPIV3/EboGP given 4 weeks apart. ELISA assay of EBOV-specific serum IgG and IgA showed that the level of EBOV-specific serum antibodies following the first dose was reduced 10-15 fold compared to the response in control animals that were HPIV3-nave. However, the serum antibody responses following the second dose were indistinguishable in HPIV3-immune versus HPIV3-naive animals. Thus, an HPIV3-based vector was substantially immunogenic even in the face of strong pre-existing immunity to the vector. Next, we deleted the F and HN genes from HPIV3 and replaced them with EBOV GP to create a virus, HPIV3/delF-HN/EboGP, in which GP would be the sole viral transmembrane surface protein. This virus was attenuated in vitro but eventually reached titers comparable to those of HPIV3. Following IN infection of guinea pigs, this virus was highly attenuated and completely restricted to the respiratory tract but nonetheless was highly immunogenic. A single IN dose provided complete protection of guinea pigs against an otherwise lethal challenge of guinea pig-adapted EBOV. Lacking the HPIV3 neutralization antigens, HPIV3/delF-HN/EboGP was insensitive to neutralization by HPIV3-specific antibodies in vitro. In addition, there was no significant difference in its immunogenicity in guinea pigs that were HPIV3-naive versus HPIV3-immune. Thus, HPIV3/delF-HN/EboGP provides an alternative to HPIV3/EboGP that is very highly attenuated, is insensitive to HPIV3-neutralizing antibodies, and nonetheless is nearly as immunogenic. We also are investigating the use of the avian Newcastle disease virus (NDV) as a human vaccine vector. NDV is antigenically distinct from common human pathogens and thus should not be affected by pre-existing immunity. In addition, there is anecdotal evidence that NDV is highly restricted in humans and does not cause significant disease. We confirmed that NDV is very highly attenuated following IN/IT inoculation of rhesus monkeys and AGM. We found that both low-virulence (lentogenic) and intermediate-virulence (mesogenic) strains replicated to similar low titers in non-human primates, suggesting that either backbone should be suitable for human vaccine purposes. Despite the high level of attenuation, which would be predictive of a high level of vaccine safety, expressed foreign proteins were moderately-to-highly immunogenic. For example, AGM that were immunized IN and IT with two doses of NDV expressing the SARS S protein developed a high titer of SARS-neutralizing serum antibodies and were strongly protected against challenge with a high dose of SARS. Another NDV was engineered to express the hemagglutinin HA glycoprotein of highly pathogenic avian H5N1 influenza virus (HPAIV) (NDV-HA). The NDV-HA virus was highly attenuated in AGM as well as in eggs and chickens. In AGM, two doses of NDV-HA induced a substantial titer of HPAIV-neutralizing serum antibodies;in addition, a substantial respiratory mucosal IgA response was induced following one and two doses, which would be particularly important in controlling a respiratory pathogen. We established a challenge model using AGM and showed that two doses of NDV-HA conferred essentially complete protection against challenge with a high dose (7.2 log10 PFU) of HPAIV. The high level of restriction of HPAIV challenge virus was established by assay of nasal swabs and tracheal lavages for virus by infectious virus assay and RT-PCR, by direct assay for infectious virus in harvested tissue, by immunohistochemical analysis of harvested tissue, and by profiling challenge-induced pulmonary host gene expression. Replacement of the polybasic cleavage site of the HA insert with the monobasic site from a low pathogenicity strain, an expedient designed to preclude any possibility of introduction of the polybasic site into circulating viruses by genetic exchange, resulted in improved immunogenicity and protective efficacy in this small study. In addition, immunization with NDV expressing the other major HPAIV surface antigen, the neuraminidase (NA) protein, also was highly immunogenic and protective. This was somewhat surprising, since the NA protein had not been considered to be a potent neutralization or protective antigen. These results showed that the modified HA gene and the NA gene are the genes of choice for inclusion in a vectored vaccine for human use. IN administration would be feasible in humans, but IT administration would not. We evaluated IN administration of the NDV construct expressing the SARS S protein and found it was not very immunogenic or protective, presumably because of insufficient vector replication in the nasal passages. Whether or not this will be predictive of replicative capability and immunogenicity in humans is unclear and can only be determined by administration to human volunteers. In the meantime, we explored an additional method of administration, namely by aerosol generated using a nebulizer. This method has been successfully and safely used in large-scale immunization against measles virus. This method of administration proved to be immunogenic and highly protective in AGM, providing a clinically relevant alternative to IN administration. In summary, NDV has considerable potential for further development as a highly attenuated vector for human vaccine use. NDV represents serotype 1 of the avian paramyxoviruses (APMV). There are 8 other serotypes, namely APMV2-9. We have initiated antigenic and sequence analysis of these as a prelude to their evaluation for attenuation and safety in non-human primates as potential vectors. Complete sequences have been determined for representatives of APMV2, 3, 4, 7, 8, and 9. In some cases, complete sequences are available for more than one strain of a serotype, namely APMV2 (2 strains), APMV3 (2 strains), and APMV8 (2 strains). Sequences already were available for two strains of APMV6: we have analyzed 2 more in addition. Also, sequencing of APMV5 is almost complete. The purpose in analyzing multiple strains is to investigate genetic diversity suggested by observed phenotypic and/or genetic diversity